Questions tagged [reinforcement-learning]

For questions related to learning controlled by external positive reinforcement or negative feedback signal or both, where learning and use of what has been thus far learned occur concurrently.

Filter by
Sorted by
Tagged with
1
vote
0answers
157 views

Policy gradient loss for neural network training

Say i want to train a neural network with 10 classes as outputs and use categorical_cross_entropy as a loss function in keras. This will try to fit the training ...
2
votes
1answer
110 views

Intuition behind $\gamma$-discounted state frequency

At the appendix A of paper "near-optimal representation learning for hierarchical reinforcement learning", the authors express the $\gamma$-discounted state visitation frequency $d$ of policy $\pi$ as ...
2
votes
1answer
39 views

Are successive actions independent?

The proof of the consistency of the per-decision importance sampling estimator assumes the independence of $$\frac{\pi(A_t|S_t)}{b(A_t|S_t)}R_{t+1}\quad\text{ and }\quad \prod_{k=t+1}^{T-1}\frac{\pi(...
7
votes
1answer
143 views

Are there reinforcement learning algorithms that scale to large problems?

Given a large problem, value iteration and other table based approaches seem to require too many iterations before they start to converge. Are there other reinforcement learning approaches that better ...
1
vote
0answers
29 views

Deciding the rewards for different actions in Pong for a DQN agent

I am attempting to implement an agent that learns to play in the Pong environment, the environment was created in PyGame and I return the pixel data and score at each frame. I use a CNN to take a ...
2
votes
0answers
184 views

Difficulty understanding Monte Carlo policy evaluation (state-value) for gridworld

I've been trying to read Sutton & Barto book chapter 5.1, but I'm still a bit confused about the procedure of using Monte Carlo policy evaluation (p.92), and now I just cant proceed anymore coding ...
8
votes
2answers
108 views

How can alpha zero learn if the tree search stops and restarts before finishing a game?

I am trying to understand how alpha zero works, but there is one point that I have problems understanding, even after reading several different explanations. As I understand it (see for example https:/...
1
vote
2answers
66 views

How do map providers like Google calculate the distance between two coordinates and find turn by turn directions?

I have searched on how Google or any map provider calculates distance between two coordinates. The closest I could find is Haversine formula. If I draw a straight line between two points, then ...
1
vote
1answer
196 views

How can the $\lambda$-return be defined recursively?

The $\lambda$-return is defined as $$G_t^\lambda = (1-\lambda)\sum_{n=1}^\infty \lambda^{n-1}G_{t:t+n}$$ where $$G_{t:t+n} = R_{t+1}+\gamma R_{t+2}+\dots +\gamma^{n-1}R_{t+n} + \gamma^n\hat{v}(S_{t+n})...
5
votes
2answers
243 views

Why don't people use projected Bellman error with deep neural networks?

Projected Bellman error has shown to be stable with linear function approximation. The technique is not at all new. I can only wonder why this technique is not adopted to use with non-linear function ...
3
votes
0answers
45 views

Deep Q-Learning agent poor performing actions. Need help optimizing

I'm trying to make deep q-learning agent from https://keon.io/deep-q-learning My environment looks like this: https://imgur.com/a/OnbiCtV As you can see my agent is a circle and there is one gray ...
1
vote
0answers
106 views

Actor-critic algorithm using gaussian Radial Basis Function, Local Linear Regression and shallow Neural Network

I'm attempting to implement the actor-critic algorithm on Matlab using Radial Basis Function, Local Linear Regression, and shallow Neural Network for inverted pendulum system. the state space and the ...
3
votes
0answers
61 views

Feature Selection using Monte Carlo Tree Search

I'm trying to tackle the problem of feature selection as an RL problem, inspired by the paper Feature Selection as a One-Player Game. I know Monte-Carlo tree search (MCTS) is hardly RL. So, I used ...
2
votes
1answer
250 views

Bachelor thesis in reinforcement learning

I've decided to make my bachelor thesis in RL. I am currently struggling in finding a proper problematic. I am interested in multi-agent RL with the dilemma between selfishness and cooperation. I ...
3
votes
0answers
142 views

What can be considered a deep recurrent neural network?

In the paper Deep Recurrent Q-Learning for Partially Observable MDPs, the DRQN is described as DQN with the first post-convolutional fully-connected layer replaced by a recurrent LSTM. I have DQN ...
5
votes
2answers
189 views

Can DQN perform better than Double DQN?

I'm training both kind of agents against an environment but DQN performs significantly better than Double DQN. As I've saw here, Double DQN use to perform better than DQN. Am I doing something wrong ...
4
votes
0answers
38 views

What research has been done on learning non-Markovian reward functions?

Recently, some work has been done planning and learning in Non-Markovian Decision Processes, that is, decision-making with temporally extended rewards. In these settings, a particular reward is ...
12
votes
3answers
738 views

Why doesn't Q-learning converge when using function approximation?

The tabular Q-learning algorithm is guaranteed to find the optimal $Q$ function, $Q^*$, provided the following conditions (the Robbins-Monro conditions) regarding the learning rate are satisfied $\...
3
votes
1answer
122 views

Understanding the n-step off-policy SARSA update

In Sutton & Barto's book (2nd ed) page 149, there is the equation 7.11 I am having a hard time understanding this equation. I would have thought that we should be moving $Q$ towards $G$, where $...
2
votes
1answer
72 views

What is the motivation behind using a deterministic policy?

What is the motivation behind using a deterministic policy? Given that the environment is uncertain, it seems stochastic policy makes more sense.
3
votes
2answers
2k views

How large should the replay buffer be?

I'm learning DDPG algorithm by following the following link: Open AI Spinning Up document on DDPG, where it is written In order for the algorithm to have stable behavior, the replay buffer should ...
2
votes
1answer
24 views

Inform policy learning of environment constants

Policy learning refers to mapping an agent state onto an action to maximize reward. A linear policy, such as the one used in the Augmented Random Search paper, refers to learning a linear mapping ...
1
vote
1answer
43 views

IQN bellman target: using Z vs using Q

IQN paper (https://arxiv.org/abs/1806.06923) uses distributional bellman target: $$ \delta^{\tau,\tau'}_t = r_t + \gamma Z_{\tau'}(x_{t+1}, \pi_{\beta}(x_{t+1})) - Z_{\tau}(x_t, a_t) $$ And optimizes: ...
3
votes
0answers
38 views

Does everyone still use discount rates?

In Section 10.4 of Sutton and Barto's RL book, they argue that the discount rate $\gamma$ has no effect in continuing settings. They show (at least for one objective function) that the average of the ...
5
votes
1answer
1k views

Can Q-learning be used in a POMDP?

Can Q-learning (and SARSA) be directly used in a Partially Observable Markov Decision Process (POMDP)? If not, why not? My intuition is that the policies learned will be terrible because of partial ...
1
vote
1answer
57 views

DQN Q-values are static

I am working on a DDQN with 5 LSTM layers and 3 actions as output and state space of 21 features. I am dividing the dataset into episodes of 720 timesteps, for each episode the agent acts greedily for ...
2
votes
1answer
161 views

What is the internal state of a Simple Neural Attentive Meta-Learner(SNAIL)?

In the paper A Simple Neural Attentive Meta-Learner, the authors mentioned right before Section 3.1: we preserve the internal state of a SNAIL across episode boundaries, which allows it to have ...
2
votes
1answer
177 views

What is the difference between A2C and running an agent in an environment vector?

I've implemented A2C. I'm now wondering why would we have multiple actors walk around the environment and gather rewards, why not just have a single agent run in an environment vector? I personally ...
5
votes
2answers
222 views

Is reinforcement learning using shallow neural networks still deep reinforcement learning?

Often times I see the term deep reinforcement learning to refer to RL algorithms that use neural networks, regardless of whether or not the networks are deep. For example, PPO (https://arxiv.org/pdf/...
2
votes
1answer
42 views

Can the normalization factor for the belief state update be zero?

In order to update the belief state in a POMDP, the following formula is used: $$b'(s')=\frac{O(a, s', z) \sum_{s\in S} b(s)T(s, a, s')}{\mathbb{P}(z \mid b, a)}$$ where $s$ is a specific state in ...
0
votes
0answers
222 views

DQN Q-mean values converge negatively

I'm trying to implement my own DQN. So far I think my code is good, but my Q-values (I'm getting the mean of all the values for every episode) tends to converge near-zero but negatively. It is normal? ...
5
votes
2answers
372 views

Is REINFORCE the same as 'vanilla policy gradient'?

I don't know what people mean by 'vanilla policy gradient', but what comes to mind is REINFORCE, which is the simplest policy gradient algorithm I can think of. Is this an accurate statement? By ...
1
vote
1answer
58 views

How to build a DQN agent which can be trained through interactive learning?

I am trying to create a chatbot whose dialogue policy model will be trained through reinforcement learning. Dialogue Policy is responsible for selecting the action to take based on the given state of ...
1
vote
0answers
69 views

Do we need to reset the DQN network after every episode?

I was going through this implementation of Reinforcement learning where model is being trained to manage the number of bikes at a station. Here, line 78 represents the loop over all episodes (if I ...
1
vote
0answers
73 views

Reinforcement Learning with limited number of episodes

I try to implement RL to a case something like this: This game consist of several rounds. Every round the players need to generate a maze that consists of rooms. There are around 1000 different ...
2
votes
1answer
191 views

Regarding the output layer's activation function for continuous action space problems

I'm interested in building a (deep) RL agent for solving a continuous problem (which splits something into portions). In all examples I've seen so far, e.g., solving the continuous lunar lander, ...
3
votes
1answer
49 views

Expressing Arbitrary Reward Functions as Potential-Based Advice (PBA)

I am trying to reproduce the results for the simple grid-world environment in [1]. But it turns out that using a dynamically learned PBA makes the performance worse and I cannot obtain the results ...
3
votes
1answer
79 views

Why Q2 is a more or less independant estimate in Twin Delayed DDPG (TD3)?

Twin Delayed Deep Deterministic (TD3) policy gradient is inspired by both double Q-learning and double DQN. In double Q-learning, I understand that Q1 and Q2 are independent because they are trained ...
3
votes
1answer
218 views

What are the differences between the DQN variants?

There are several variants of the DQN model. For example, double DQN, duelling DQN, prioritized DQN, distributed prioritized DQN, episodic memory DQN, asynchronous n-step DQN and multiple DQN. What ...
6
votes
1answer
3k views

What is the difference between reinforcement learning and optimal control?

Coming from a process (optimal) control background, I have begun studying the field of deep reinforcement learning. Sutton & Barto (2015) state that particularly important (to the writing of ...
2
votes
2answers
115 views

Can Reinforcement Learning solve problems, where certain elements in the environement are randomly located?

I want to solve a problem using Reinforcement Learning on a 20x20 board. An agent (a mouse) has to get the highest possible rewards as fast as possible by collecting cheese, which there are 10 in ...
3
votes
2answers
434 views

Reinforcement Learning with long term rewards and fixed states and actions

I have read a lot about RL algorithms, that update the action-value function at each step with the currently gained reward. The requirement here is, that the reward is obtained after each step. I ...
1
vote
2answers
82 views

Running 2 NEAT nets on the same observations

So i have been playing around with neat-python. I made a program, applying neat, to play pinball on the Atari 2600. The code for that can be found in the file ...
3
votes
1answer
39 views

How agent's reasoning skills can improve its reinforcement learning?

Agent can have reasoning skills (prediction, taking calculated guesses, etc.) and those skills can help reinforcement learning of this agent. Of course, reinforcement learning itself can help to ...
2
votes
0answers
34 views

Hindsight Experience Replay with multiple goals

What if there are multiple goals? For example, let's consider Bit-flipping environment as described in the paper HER with one small change: Now, goal is not some specific configuration, but let's say ...
5
votes
1answer
94 views

Reinforcement Learning with more actions than states

I have read a lot about RL recently. As far as I understood, most RL applications have much more states than there are actions to choose from. I am thinking about using RL for a problem where I have ...
2
votes
0answers
61 views

Reinforcement learning for inventory management with dynamic changes to available products

Consider a shop owner who has to deal with having to buy for one week from a different supplier with several different brands. Another week a brand is removed or added from the market. Yet another ...
1
vote
1answer
214 views

Importance of starting state and player in RL for Tic Tac Toe

So I am simulating a Tic Tac Toe game with a human opponent. The way the RL trains is through policy/value iterations for a fixed number of iterations all specified by the user. Now whether the human ...
0
votes
1answer
40 views

Why don't we decorrelate transitions for policy-based data?

I'm implementing PPO myself strictly follow the steps: sample transitions randomly shuffle the sampled transitions compute gradients and update networks using the sampled transitions drop transitions ...
2
votes
0answers
149 views

Is Reinforcement Learning the future of Natural Language Processing?

I was reading about the grounding problem after seeing it mentioned in another answer today. The article states that, in order to avoid the "infinite regress" of defining all words with other words, ...

1
12 13
14
15 16
22